Pure electric commercial vehicles are extensively utilized under diverse driving conditions, each imposing distinct demands on the drive system. This variability complicates the design of power systems for commercial vehicles, making the optimal matching of drivetrain parameters crucial. To maximize vehicle performance and efficiency, this study integrates numerical statistical methods during the parameter-matching process of the vehicle's drivetrain system. By analyzing the numerical characteristics of driving conditions and incorporating the determined energy distribution and motor high-frequency operating ranges, a multi-objective optimization of the transmission system is performed using a combined Cruise and iSIGHT simulation with an improved Non-dominated Sorting Genetic Algorithm II. The results reveal that drivetrain parameters matched using numerical statistical methods surpass traditional methods. Post-optimization, the pure electric commercial vehicle not only meets power requirements but also achieves a 2.85% improvement in economic performance under Chinese Heavy Commercial Vehicle Driving Conditions cycle conditions and a 1.86% increase in efficiency at the design target speed of 88.5 km/h.